Risk parity strategies have had quite a month. 'Risk on / Risk off' market dynamics of the last several years have been given up for a strong 'everything off' trade and for a couple days there, the mantra of the market seemed to be: "The beatings will continue until morale improves."

As we've written about previously, the trouble with risk parity is the details of the implementation.  With assets across the board flashing red, balancing risk factors hasn't made much of a difference -- especially when risk is measured via correlation and a symmetric measure like volatility.  But while we lauded Bridgewater for their implementation, it seems that even their fund is feeling under the weather.  Other retail risk-parity strategies haven't held up well either this month.

Risk Parity

Being long an asset has a degree of similarity to selling insurance: you collect a return premium for exposing yourself to certain risks (which is why people sometimes explain markets as a vehicle for packaging and transferring risk).  Investment managers seek to identify and invest in assets paying outsized excess returns for their risks, whether because the magnitude of the risks or the probability of the risks are over-estimated by the market.  The trouble, of course, is that when these risks are realized, they typically wipe out the premiums we've collected (and more!).  So, just as an insurance company wouldn't only sell a single type of insurance in a single geographic region, we wouldn't sell insurance on a single risk for a single asset class in our portfolio.  If we can construct a portfolio with equally balanced exposure to independent risks, we can dramatically reduce our risk of ruin.

In a way, that's what unconstrained mean-variance optimization does: correlation is used to create independent long/short "principal portfolios" that are then levered to have equal volatility ("risk allocation") and weighted based on their relative, levered expected excess return.  Long-only portfolios achieve equal risk allocations by investing more in  "low risk" asset classes.  This has the effect of dampening portfolio volatility.  That's where risk-parity comes in, using leverage to increase asset-level volatility to a target level so that overall portfolio volatility doesn't become dampened in the diversification process.

The trouble with naive implementations of risk parity is that correlation and volatility are not necessarily enough to capture more transient risk factors like duration, liquidity, and forcible deleveraging.  As economic dynamics change, the probability of certain risks being realized changes and markets attempt to price these changes.

In the last month, it seems that risk relationships have flip-flopped over-night.  Falling rates and low correlations to equities made bonds a very attractive asset class over the previous several decades.  A swift rise in interest rates made government bonds, a previously a low risk, steady positive return  asset into a higher risk, negative return asset.  Couple this with realized positive correlations to equities and you have a recipe for losses.

The question is, have we wandered into a potentially massively levered bond bet at the precipice of a bull market in bonds?  Only time will tell.

Then again, maybe this is all overblown and premature.  Even a very naive implementation, ignoring correlation and using inverse volatility weights between SPY and AGG has seen worse sell-offs.  
naive risk parity equity curve dd

Corey is co-founder and Chief Investment Officer of Newfound Research, a quantitative asset manager offering a suite of separately managed accounts and mutual funds. At Newfound, Corey is responsible for portfolio management, investment research, strategy development, and communication of the firm's views to clients.

Prior to offering asset management services, Newfound licensed research from the quantitative investment models developed by Corey. At peak, this research helped steer the tactical allocation decisions for upwards of $10bn.

Corey is a frequent speaker on industry panels and contributes to ETF.com, ETF Trends, and Forbes.com’s Great Speculations blog. He was named a 2014 ETF All Star by ETF.com.

Corey holds a Master of Science in Computational Finance from Carnegie Mellon University and a Bachelor of Science in Computer Science, cum laude, from Cornell University.

You can connect with Corey on LinkedIn or Twitter.